This proposed study of urban Black and Latina women has two major aims. The first is to examine the longitudinal pathways to HIV risk behaviors (unprotected intercourse, multiple/concurrent partners, having a high-risk partner) focusing on the integration of structural/contextual with interpersonal and individual risk factors as predictors. Drug use will be examined as a behavioral risk factor mediating the relationship between structural risks and HIV-risk behaviors. The proposal is informed by Family Interactional Theory and the Theory of Gender and Power, as it applies to HIV-risk behaviors among women of color. The proposed research uses a life-course perspective to examine this critical issue by focusing on both antecedent and concurrent predictors of engagement in HIV-risk behaviors among African American and Puerto Rican women in their thirties. Second, the research will examine the ability of resource factors (i.e., ethnic pride, problem-focused coping, social support) to mitigate structural risk factors (e.g., neighborhood disadvantage) for HIV-risk behaviors. This grant proposal will capitalize on a wealth of data collected at key developmental stages over a period of 20 years as part of the Harlem Longitudinal Development Study (HLDS). Identification of the risk factors for HIV-risk behaviors and the pathways through which they operate will have important implications for public health. Findings from this research can be integrated into prevention and intervention programs aimed at reducing the transmission of HIV and other STIs among women disproportionately at risk. The sample for this study will consist of 450 women who were previously seen in adolescence (T1, T2), emerging adulthood (T3), and young adulthood (T4). At T5, face-to-face interviews will be conducted with the participants. Psychometrically sound scales will be developed from the surveys.